Columnar storage, clearly explained
Our partner Exasol explains why column-oriented storage works so well for analytical workloads.
New to the concept of columnar storage? Our partner Exasol has published a clear explanation.
Analytical queries usually touch only a few columns, across a very large number of records. A columnar database stores the values of one column together, so the engine reads only what the query actually needs. That means less I/O, less memory and less data movement, and it enables compression and fast aggregations. That's why it suits BI, reporting, data warehousing and dashboards so well.
Exasol rightly points out that storage is not the whole story: query design, data model, hardware and the database implementation together determine performance.
The benefits for analytical workloads are huge, and immediately visible in a live demo. Interested? Get in touch.